This study systematically reviews 257 peer-reviewed articles indexed in the Scopus database and published between 2005 and 2025 to characterise research trends, methodological paradigms, geographic distribution, and indicator usage in urban noise and soundscape mapping. The temporal analysis shows that the topic first emerges in 2005 with pioneering studies from Nigeria and Brazil, followed by a pronounced rise in publication activity after 2016, driven primarily by developments in China, India, Brazil, and several Southeast Asian countries. Methodologically, the field has shifted from diagnostic measurements toward integrated systems incorporating in-situ monitoring, GIS, simulation tools, and, more recently, machine-learning methods. Experiential and soundscape-based studies remain limited but are gradually expanding, whereas policy-oriented and health-integrated assessments are still rare. Overall, the review identifies four persistent challenges: geographic and thematic imbalance, limited integration of perceptual and socio-spatial dimensions, weak policy translation, and a notable absence of health and wellbeing focused, human-centered mapping. Addressing these gaps is essential for producing robust, comparable, and policy-relevant acoustic evidence capable of informing urban planning and environmental-health strategies across the Global South.
Environmental noise assessments in Europe are mostly based on the infrastructure categories prescribed by the Environmental Noise Directive (END), which may not represent the full exposure conditions experienced by populations and ecosystems. This study quantifies how that strategic scope affects exposure estimates by comparing END-restricted road traffic noise simulations with a complementary modelling configuration that includes all road categories. A regionally validated noise model was implemented over the Region of Murcia (Spain) using high-resolution acoustic propagation and large-scale parallel computation. Results show that END-based assessments capture only a fraction of the affected territory and population: the area where noise exceeds moderate levels expands by more than a factor of three when all roads are included, and the number of residents exposed above relevant health-based thresholds nearly doubles. Protected areas also display substantial additional exposure, revealing diffuse noise propagation into ecological sites that remains undetected under the regulatory scope. These findings show that END-based indicators, while appropriate for strategic reporting, capture only part of the broader exposure field because large portions of the transport network remain outside the mapped domain. The results therefore support the use of complementary full-network modelling when the objective extends beyond END reporting to exposure assessment, health-oriented analysis or environmental planning, and indicate that such simulations are tractable with current high-performance computing infrastructures.
Automated bird species classification is a critical component of biodiversity assessment and ecological monitoring; however, conventional unimodal systems often suffer significant performance degradation in noisy and visually challenging environments. This paper presents AVFusionNet, a novel noise-aware multimodal deep learning framework that integrates visual and acoustic cues for robust bird species recognition. The proposed system employs dual convolutional neural network (CNN)-based feature extractors to learn spatial representations from bird images and temporal-spectral representations from log-Mel spectrograms of vocalizations. A key contribution is an adaptive feature fusion strategy that dynamically prioritizes the more reliable modality under adverse environmental conditions, thereby enhancing robustness and decision stability. A custom multimodal dataset comprising four bird species - Chicken, Crow, Duck, and Parrot - was constructed and evaluated under 10 controlled scenarios involving varying levels of image degradation and acoustic noise. Experimental results show that the proposed model achieves 98.73 % classification accuracy under ideal conditions and maintains performance above 80 % across all degraded scenarios, with accuracy ranging from 82.50 to 93.75 % under visual degradation, 81.25-98.73 % under acoustic noise, and 80-95 % under simultaneous multimodal degradation. These results demonstrate that AVFusionNet provides robust and consistent performance, significantly outperforming conventional CNN, support vector machine (SVM), and random forest approaches, and is well suited for real-time ecological monitoring and intelligent wildlife surveillance applications.
Historically, environmental noise measurements have been performed at heights of 1.2–1.5 m, until the European Union defined 4 m as the reference height for strategic noise maps. This has led to the question of whether measurements at one height or the other make a difference; the question arises because pedestrians’ exposure could be underestimated by sound pressure levels measured at 4 m above ground level. This paper presents an analysis of 243 simultaneous measurements at 1.5 m and 4 m above ground level, performed in five cities in Spain and Uruguay with different populations, in locations with distinct characteristics. The objective was to compare their signs and absolute values, and to determine whether they can be considered statistically comparable with a 95 % confidence level. For this purpose, statistical tests were performed. The results of the Wilcoxon pairwise differences test showed that 80 % of the series were not statistically comparable. In light of this fact, environmental sound pressure level measurements should be conducted at the height of interest, since the phenomena described by measurements at different heights don’t describe in the same way the same acoustic phenomena.
Traffic noise is the common component at road intersections in the outdoor environment, and the sounds of traffic light signals are designed to distinguish signal transition between red and green. However, the sounds of traffic light signals are often masked by high levels of vehicle noise produced by high traffic flow. Therefore, this study aims to investigate the acoustic effects of traffic light signals and noise on sound sensitivity in the outdoor environment. A set of psychoacoustic indicators were applied to assess the traffic noise consisting of traffic light signals and vehicle noise at road intersections in Hong Kong. Furthermore, the subjective listening tests evaluated the acoustic sensitivity and perceptions. The results of psychoacoustic time-varying curves revealed distinct characteristics of traffic light signals, with fluctuation strength differing significantly between red and green signals. Moreover, the subjective perceptions of rapid tempo and regularity were significantly correlated with the measured fluctuation strength and rhythmic events per minute (REPM) during the signal transition. These findings will be the insight into the acoustic interactions between various sound sources and their perceptual implications in the outdoor environment.
This study evaluates the impact of road traffic noise on roadside urban schools by adopting a dual methodological approach that combines quantitative noise monitoring and mapping with a socio-acoustical survey. 10 schools located near busy roads with varying noise exposure levels were examined. Noise levels were measured both inside classrooms and along adjacent roadways, and key acoustic indices including Traffic Noise Index (TNI), Noise Climate (NC), and Noise Pollution Level (Lnp) were computed. Noise propagation maps were generated to visualize extent and spatial distribution of noise exposure around each school. Results show that observed noise levels frequently exceeded the World Health Organization (WHO) recommended limits, with exceedances ranging from 15 % in S5 to 107 % in S6. The socio-acoustical survey further highlighted the perceptual and psychological impacts of noise, including reduced speech intelligibility and increased emotional stress. Approximately 80 % of students reported frequent disturbance from external noise, 65 % experienced loss of concentration, and 50 % reported feelings of annoyance or fatigue. While the findings provide substantial evidence of the adverse effects of traffic noise on roadside schools, further research is needed to deepen understanding and explore long-term implications.
Road traffic noise is a significant environmental risk factor to the human health, resulting in the necessity for accurate noise prediction models with less uncertainty. Several standards and methods are available in commercial noise modeling software to predict noise levels from road traffic noise, such as ISO 9613-2, CNOSSOS-EU, RLS 19, NMPB2008, RMG2012 and Nord2000. However, many of these standards were developed for specific national conditions and thus have limitations in wider applications. For instance, the RLS 19 standard is most suitable for the German road network, the NMPB2008 standard is tailored to French conditions, the RMG2012 is Dutch and the Nord2000 standard is predominantly applied in Northern Europe. Similarly, the use of CNOSSOS-EU is limited to European Union member states that have transposed the method into national law, as recommended by the Environmental Noise Directive (END), due to varying road surface characteristics across countries. In contrast, the ISO 9613-2 standard, entitled “Acoustics – Attenuation of sound during propagation outdoors – Part 2: Engineering method for the prediction of sound pressure levels outdoors” provides a general engineering framework for estimating noise levels from several sources, such as road traffic noise. This paper presents the results of a thorough investigation in which predicted noise levels from ISO 9613-2 and CNOSSOS-EU models were compared with measured data for a road section with relatively low traffic volume. The purpose of this study is to identify the modeling approach that delivers the most realistic predictions, offering guidance on selecting appropriate methods and techniques for similar traffic conditions. The findings contribute to a better understanding of the applicability of the evaluated methods for noise modeling of roads with low traffic volumes, highlighting their respective strengths and limitations. This comparison can support environmental authorities and acoustic consultants in selecting the most suitable prediction method for local and national assessments, and it provides a basis for improving model calibration and adaptation in future studies.
Noise pollution is a growing threat to public health in dense cities yet widely used quietness indices were calibrated in European contexts and transfer poorly to Asian megacities. This study proposed the Adaptive Quietness Suitability Index (AQSI), a refined, validated framework that enhances sensitivity, accuracy, and transferability in mixed-use urban environments. The AQSI advances three elements: (i) dynamic impact-zone delineation via physics-based sound propagation rather than fixed buffers, (ii) alignment with locally regulated noise limits and control zones, and (iii) explicit temporal stratification (daytime, evening, nighttime, all-day) to capture diurnal variability. We validated the AQSI using monitoring data from 1997 to 2024 in New Taipei City, Taiwan, and compared it to the original QSI through regression and spatial analyses. The AQSI showed stronger agreement with measured sound levels than the original QSI (Pearson’s r = −0.238 vs. −0.202), representing a 39 % improvement in explained variance (R 2 = 0.057 vs. 0.041). It eliminated extreme-value saturation (0 or 1 in 86.2 % of cells) and reduced it to 0 %, yielding continuous gradients that better resolved intermediate environments. Nighttime emerged as the critical period, with the steepest negative coefficient (−0.0142, p < 0.001) and the highest adjusted R 2 (0.16), highlighting the need for nocturnal noise management. By integrating operational mapping with soundscape-aware considerations at a 50 m resolution, the AQSI provides a transparent, context-sensitive tool for urban planning, regulatory compliance, and targeted mitigation in diverse metropolitan settings.
Urban traffic noise is associated with the health and living environment quality of residents. As urbanization and population density continually increase, it is vital to understand and predict the impact of urban design behavior on urban traffic noise. Despite the current progress has been made in modeling traffic noise using limited land use types, understanding the complex relationship between various land uses and traffic noise remains challenging for stakeholders. This study used generative adversarial networks with Hong Kong one-hour peak traffic noise map to predict urban traffic noise. The applicability of the training model was evaluated through accuracy analysis and validation. The validated model was used to generate the predicted noise map in multiple scenario experiments by adjusting controlled variables. This approach explores how land use changes effect the noise level, with scenario experiments highlighting both effective strategies and areas requiring further validation.
Uruguay is a small country in Latin America. It has 178.500 km2 and approximately 3,400,000 inhabitants. Its environmental legislation is still incomplete; for example, there has been a national act about noise pollution since 2004 but it has never been regulated. Thus, no national regulation on noise but only departmental Ordinances in each of its 19 Departments; in practice, 19 different regulations coexist on such a small surface. Noise maps are not mandatory neither in Uruguay nor in any of its Departments. The Research Group on Environmental Noise at Universidad de la República has developed a research project that seeks the best practical methodology to build noise maps through manual measurements. The fieldwork included the determination of the stabilization time of noise measurements, the comparison between long- and short-time measurements, the comparison between measurements taken at 1.20 m and 3.50 m, and the obtention of a national curve of highly annoyed people (% HA) with a basis in the field measurements and simultaneous survey carried out on site. In this article, we present the results of these works and the proposed methodology for building noise maps throughout the country.
Europe is acting to fight noise pollution. The Environmental Noise Directive (2002/49/EC) requires EU Member States to determine the exposure to environmental noise through strategic noise mapping, and elaborate action plans to reduce noise pollution. Road traffic noise is a common environmental noise source; henceforth, EU countries are obliged to produce strategic noise maps for all major roads, railways, airports, and agglomerations, on a 5-year basis. These noise maps are used by national competent authorities to identify priorities for action planning and by the European Commission to globally assess noise exposure across the EU. A thorough investigation is conducted in this article to assess how different road surface types affect road traffic noise levels in selected EU member states which are incorporating CNOSSOS-EU into their national law. It has been done by comparing the nationally published noise data to those published by CNOSSOS-EU in 2021 for various vehicle categories by obtaining the rolling and propulsion noise for each road surface type while ignoring other factors. The aim of this study is to address the deficiency in the assessment and show a comparison between the noise generated from different surface types which can potentially enhance the effectiveness of strategic noise mapping.
Determining the land cover (LC) data requirements used as input to noise simulations is essential for planning sustainable urban densifications. This study examines how different LC datasets influence simulated environmental noise levels of road traffic using Nord2000 in an urban area of 1 km2 in southern Sweden. Four LC datasets were used. The first dataset was based on satellite data (spatial resolution 10 m) combined with various other datasets implementing an LC classification algorithm prioritizing vegetation. The second dataset was created by applying an LC majority priority rule over every cell of the first dataset. The third dataset was produced by applying a convolutional neural network over an orthophoto (0.08 m spatial resolution), while the fourth dataset was created by manually digitizing ground surfaces over the same orthophoto also utilizing data from the municipality’s basemap. The results show that LC data impact simulated noise levels, with priority rules in LC classification algorithms having a greater effect than spatial resolution. Statistically significant differences (up to 3 dB(A)) were found when comparing the simulated noise levels generated using the vegetation-prioritizing LC dataset compared to the simulated noise levels of the other LC datasets.
Due to rapid urbanization and industrialization, noise pollution has become a growing global concern, with significant impacts on occupational and environmental health. Unlike earlier times when it received limited attention, its importance has increased due to mounting evidence of its health effects. Research on noise pollution highlights its consequences and helps identify gaps that require further exploration. This systematic review aims to compile and categorize the health effects associated with various sources of noise pollution.
Urban environments are characterized by a complex interplay of various sound sources, which significantly influence the overall soundscape quality. This study presents a methodology that combines the intermittency ratio (IR) metric for acoustic event detection with deep learning (DL) techniques for the classification of sound sources associated with these events. The aim is to provide an automated tool for detecting and categorizing polyphonic acoustic events, thereby enhancing our ability to assess and manage environmental noise. Using a dataset collected in the city center of Barcelona, our results demonstrate the effectiveness of the IR metric in successfully detecting events from diverse categories. Specifically, the IR captures the temporal variations of sound pressure levels due to significant noise events, enabling their detection but not providing information on the associated sound sources. To fill this weakness, the DL-based classification system, which uses a MobileNet convolutional neural network, shows promise in identifying foreground sound sources. Our findings highlight the potential of DL techniques to automate the classification of sound sources, providing valuable insights into the acoustic environment. The proposed methodology of combining the two above techniques represents a step forward in automating acoustic event detection and classification in urban soundscapes and providing important information to manage noise mitigation actions.
In the field of urban design assessment, increasing attention to the nonexpert verbalizations is paid to engage social participation in planning processes, especially those not limited to static visual renderings. However, little has been explored on reproducible methods to collect these vocabularies and their relationship with architectural features. This study presents a hierarchical multifactor analysis to extract the main perceptual nonexpert clusters; and a linear correlation analysis to relate them with 2D isovist measures. In an on-line experiment, participants (n=20n=20) elicited individual attributes (n=120n=120) to describe their soundscape and visual perception and compared recorded urban environments (n=8n=8). The results show that a percentage of nonexpert audio attributes correlate as well as the visual ones with isovist metrics.
Abstract This article systematically reviews research on noise pollution monitoring conducted over the past 23 years at various bus transit terminals located worldwide. About 18 articles were identified using PRISMA method and were evaluated to provide summary of prior research work to examine accuracy, authenticity, and reliability of noise monitoring results with respect to chosen methodology and extent of noise pollution at bus transit terminals. It examines important indicators of noise pollution and the analysis parameters such as noise sampling, noise descriptors, processing of acquired data, noise mapping, etc., and compares it with the regulations and standard guidelines notably ISO 1996-2:2017 and American National Standards Institute/ASA S12.18-1994 (R2009) and their prior versions aiming to identify research gaps. The studies have primarily focused on noise monitoring, revealing widespread excessive noise pollution exceeding permissible levels at bus terminals globally. This article underscores significant research deficiencies in noise pollution monitoring at bus terminals, emphasizing the challenge of conducting quantitative meta-analyses and statistical comparisons due to variations in parameters and qualities. Noise pollution standards are breached in all terminals covered in the identified literature; hence, noise mitigation measures must be implemented at these bus terminals. The study suggests that noise monitoring must be carefully devised with respect to individual site operations and noise sources and in compliance with standard guidelines to improve the accuracy of the results. There is a need for uniform guidelines that can be followed globally for environmental noise monitoring as there are only a few countries that have guidelines for noise monitoring. The outcomes of this research will be helpful in guiding noise monitoring, mapping, and mitigation strategies as well as designing transit terminals to improve overall acoustical ambiance for more passenger footfall for sustainable transportation.
Environmental noise directive requires Member States to produce and periodically update strategic noise maps for agglomerates in order to evaluate citizens’ noise exposure before the action plan. Its ultimate purpose is to mitigate the highest or most harmful noise levels. Available tools do not provide a point-to-point indication of the predominant sources to be addressed by intervention on the sources by individual owners. Recently, noise source predominance maps were developed to show the predominant source at each point of the calculation grid of the strategic noise map by means of polygons and colors. Intensity noise source predominance maps were also developed to add the visualization of noise exposure levels by coloring the polygons according to a color scale. This work investigates their applicability in agglomerates and defines an additional indicator with the purpose of connecting the high levels of citizen exposure to its predominant source. The new approach provides an aid to administrations in identifying areas where singular noise mitigations would be truly effective in improving the citizen’s quality of life by ensuring a noticeable decrease in total noise while reducing exposure to the specific source.
This study proposed a method of obtaining the type and quantity of equipment in factories by inquiring environmental impact assessment reports, which greatly improves the efficiency of gathering factory information. Thereafter, by combining on-site measurement and numerical modelling, the noise maps of an automobile industrial area were constructed. The exposed population under different noise levels were evaluated using the noise maps. The results indicated that noise pollution at nighttime in the study area was more severe than that during daytime, with 523 people (1.08%) and 1,357 people (2.81%) exposed to excessive noise levels during daytime and nighttime, respectively. In addition, this study also constructed a low-frequency industrial noise map. The methods and results of the present study can provide novel technical path for construction and analysis of large-scale industrial noise map.
The hospital soundscape is known for high noise levels and a perception of chaos, leading to concerns about its impact on patients, families, professionals, and other hospital staff. This study investigates the relationship between sound, Annoyance, and sleep quality in a multi-patient neurology ward. A mixed-methods approach was employed. Interviews were conducted with medical staff (n = 7) to understand their experiences with sound. Questionnaires and sleep tracking devices (n = 20) assessed patient sleep quality and Annoyance caused by sound events. In addition, listeners (n = 28) annotated 429 nighttime audio recordings to identify sound sources and rate Annoyance level, which we considered the key emotional descriptor for patients. Over 9,200 sound events were analysed. While snoring, a patient-generated sound dominated the nighttime soundscape and was highly rated for Annoyance, and staff-generated sounds such as speech and footsteps were found to contribute more to accumulated Annoyance due to their extended duration. This study suggests that patient sleep quality can be improved by focusing on design interventions that reduce the impact of specific sounds. These might include raising awareness among staff about activities that might produce annoying sounds and implementing strategies to mitigate their disruptive effects.
